Hydro Test Pressure Limits when Pipeline has Flanges (Big Inch versus Little Inch) 2

[auzie5]

Anybody think a NPS 4 x 4.0mmWT CL600 RFWN flange would hold at 22,337.8 kPa? Think the gasket would blow out before bending the flange?

Background:

Assume the Cold Working Pressure of a CL600 flange is 9,930 kPa (which it is in the CSA Z245.12 flange standard).

Therefore, 1.5 x CWP = 1.5 x 9,930 kPa = 14,895 kPa

1.5 x CWP (ex: 14,895 kPa) is often times used to define the maximum pressure limit during a strength test for a line with flanges in it.

For a bigger inch pipelines (ex: NPS 20 x 9.5mmWT, Gr. 448), using 1.5 x CWP isn’t that bad since a strength test at 14,895 kPa corresponds to a hoop stress equivalent to ~88.9% SMYS (at the high point). A pretty good strength test.

However, for smaller flow lines (ex: NPS 4 x 4.0mmWT, Gr. 359), using 1.5 x CWP results in a pretty crumby pressure test. I would need to bring the test section up to 22,337.8 kPa in order to develop a hoop stress equivalent to ~88.9% SMYS. And limiting the test to 1.5 x CWP (i.e. 14,895 kPa) only develops a hoop stress equivalent to 59.3% SMYS (at the high point).

Since the goal of the pressure test is to eliminate as many defects as possible, it is good practice to set the target test pressures as high as possible (ex: 85% SMYS; or higher for critical applications).

Is the general consensus to limit the test pressure to less than 1.5 x CWP for these small flow lines that have flanges, valves, etc. in them? I can test the pipeline portion at a higher pressure separate from the risers/trap assemblies; but just curious what other people are doing for these little inch lines.

Notes

MSS-SP-44 (Steel Pipeline Flanges) specifically states:

“8.2) Flange Testing - Flanges are not required to be hydrostatically tested. Flanged joints may be subjected to system hydrostatic tests at pressures not exceeding 1.5 times the 38 degC (100 deg F) rating.”​

However, I do not believe B16.5 has that same limitation. CSA Z245.12 certainly does not.

Furthermore, CSA Z662-15 Table 8.1 outlines the “Test requirements for steel piping” and limits the maximum strength pressure to be “110% of the SMYS of the pipe”.

Notice the specific wording used in Table 8.1 (i.e. “piping” in the title versus “pipe” used to limit the maximum pressure during the test).

Piping is defined as a portion of a pipeline system, consisting of pipe or pipe and components (i.e. pipe and flanges). This implies that it is permissible to take a pipeline with flanges in it up to a pressure corresponding to a hoop stress equal to 110% SMYS of the pipe. For our NPS 4 x 4.0mmWT Gr. 359 example, that mean we can take the line well past 22,337.8 kPa.

B31.4 is a little more straight forward. Para. 437.1.4 (a)(4) states, “When testing piping, in no case shall the test pressure exceed that stipulated in the standards of material specifications…for the weakest element in the system…”. However, I do not believe B16.5 explicity states that flanges can never see a pressure greater than 1.5 x CWP rating.

Thanks in advance for any input.

 

[BigInch]

When I first started designing pipelines I worked for a company that had been exclusively onshore and they had the habit of testing their pipelines to 110% SMYS, possibly because what you say, which I never believed, "The goal of the pressure test is to eliminate as many defects as possible". It's not. It is to determine the worthiness of the constructed assembly to hold the design pressure. That's all. What more is truely necessary than that. That's why I convinced that company to do tests only to the prescribed pressures. My reasoning was you'd have to be nuts to go looking for a failure to happen during a hydrotest offshore where it might cost 10x the price to repair, if it was possible to repair it at all. Test pressures are not prescribed based on obtaining a high percentage of yield strength, being rather set to a certain factor above Design Pressure. If I choose to build a pipeline to a design pressure of say 50% SMYS, the minimum required test pressure would be only 1.25 x 50% SMYS = 0.625 SMYS. Leave it at that. There are plenty of defects in any pipeline. We know that nothing is perfect. What we hope to find are only those defects that we know we wouldn't want to live with.

 

[LittleInch]

For a moment I thought you were pitting me against my brother here....

I agree with what BU+i says and that the purpose of the pressure test, IMHO, is not to eliminate as many defects as possible, but is a strength test to a suitable proportion of the design pressure, but often limited to a certain percent of the SMYS. Piping goes for 1.5, pipelines go for 1.25 because they design the pipe thinner.

ASME B 16.5 has the same or similar wording, section 2.6 Flanged joints and flanged fittings may be subjected
to system hydrostatic tests at a pressure of 1.5 times the
38°C (100°F) rating rounded off to the next higher 1 bar
(25 psi) increment. Testing at any higher pressure is
the responsibility of the user, taking into account the
requirements of the applicable code or regulation.

In terms of what the CSA code says, I think you're being obtuse. It is patently clear that the pressure test is limited by the element within any tested system to the lowest rated item.

This desire to test to 105 or 110% of SMYS arose from test undertaken on large diameter gas pipeline subject to long term frequent diurnal loadings and hence increased fatigue loading. It was shown in those instances that testing to above the SMYS essentially "blunted" small cracks which otherwise would grow over time due nto the SCF at the tip of the crack. In those case I have some sympathy for the 105 or 110% SMYS case, but given that is pipeline, the pressure even then was normally not more than the 1.5 x DP of the flanges.

In terms of smaller pipe, yes you don't test to the same SMYS, but time has shown that smaller pipes don't fail any more than others and in reality have more metal. The issue, like your other post, is that a small defect, corrosion or other loss of metal on a 4mm pipe makes a bigger impact than on an 8 or 12mm wt pipe.

Overstressing flanges, bolts and other items to the extent you suggest is not going to make your system safer, only less so IMHO.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[jte]

MSS-SP-44 (Steel Pipeline Flanges) specifically states:


“8.2) Flange Testing - Flanges are not required to be hydrostatically tested. Flanged joints may be subjected to system hydrostatic tests at pressures not exceeding 1.5 times the 38 degC (100 deg F) rating.”


However, I do not believe B16.5 has that same limitation. CSA Z245.12 certainly does not.

auzie5-

Take a peek at B16.5 paragraph 2.6. This paragraph explicitly allows testing up to 1.5* the ambient rating while not limiting the pressure. The decision to test at pressures higher than 1.5* is left to the user and the code of construction of the system. Paragraph 8.1 notes that the flanges themselves need not be pressure tested.

I'll stay out of CSA codes... I'm familiar with the ASME variety only.

 

[Gator]

"There are plenty of defects in any pipeline. We know that nothing is perfect. What we hope to find are only those defects that we know we wouldn't want to live with."

I like that. Who was it that said, "Perfection is the enemy of Good Enough"?

Piping Design Central

 

[BigInch]

Perfection is the enemy of cost controls too and, being that it exists only in the eye of God anyway, chasing it is akin to chasing the rainbow. That's why good specifications never demand perfection; i.e. every dimension should include a tolerance. This is also similar to the proverbial case of a surgeon that had to leave a bullet inside the patient simply because it was far too dangerous to take it out.

 

[zdas04]

Gator,
I don't know who said it, but I'd like to buy him a beer. "Good Enough is a red-thread theme throughout my 5-day facilities engineering course. I always try to get entire groups to come to the training together so that I have a mix of old hands and newbys and the old hands always get the concept of "good enough", but it is a struggle to get the newbys to stop shooting for perfect. They usually don't get it until the first project budget and schedule they blow.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[SNORGY]

Does it not say somewhere in B16.5 that your test pressure should not exceed 1.5 times the ceiling pressures listed in Appendix A.(x)? (I forget if x = 1 or 2).

That's the direction I see a lot of Client specifications going these days.

 

[KevinNZ]

Pressure tests are not strength tests. They are leak tests, see the wording in B31.1 and B31.3.

The pipe (and the longitudinal welds) has been pressure tested at the factory. Usually way more than 1.5 the design pressure of the system.

In using the pipe in the system you have added butt welds. A pressure test does stress these weld by much at all.

Piping is not like pressure vessels where the hydro test is used to disrupt fabrication stresses.

 

[LittleInch]

KevinNZ,

Therein lies a key difference between piping ( 31.3/31.1) and pipelines )31.4, 31.8 etc). If you read the OP closely, he or she is clearly talking about flanges on pipelines.

Yes 31.3 says the hydrostatic test is a leak test. but 31.4 refers to it as a "hydrostatic proof test" which talks about leak tests at a lower pressure.

Piping is not normally stressed to eliminate or reduce fabrication stresses and small cracks as you state, but pipelines sometimes are, especially gas pipelines in some companies view.

snorgy - No it doesn't say that anywhere. It says the rating pressure should not exceed the ceiling pressure (a1 is metric, A2 is Imperial). The testing section is 2.6 copied above vertabim. Upper limit above 1.5 is "the responsibility of the user".

We've had many posts here saying "I went 40/50/100 psi over the limit, what now??" Normally nothing. 500 psi over the limit though....

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.